Number of arrests in the UK over social media posts
Executive summary
The best available reporting indicates roughly 12,000 arrests in the UK in 2023 for communications offences that include social media posts, a figure widely cited by The Times and synthesised in Freedom House's 2025 country report (over 12,000 arrests) [1] and by summaries noting an average of about 12,000 arrests per year between 2021–2023 [2]. That headline number comes with important caveats: the datasets are based on custody and FOI returns that use broad offence codes (which cover messages by any medium), some police forces did not provide complete data, and conviction and imprisonment rates under those laws are much lower [3] [4] [5].
1. The headline figure: where “12,000” comes from and what it covers
Reporting from The Times and subsequent summaries show more than 12,000 arrests in 2023 under section 127 of the Communications Act 2003 and section 1 of the Malicious Communications Act 1988, and commentators have translated that into “over 30 arrests a day” or roughly 12,000 annually [1] [6] [2]. Those figures derive from custody data and freedom‑of‑information returns aggregated by national reporting, but they are counts of arrests recorded under those offence codes — not a clean count of arrests exclusively for social media posts — because the legal provisions cover communications by any medium, including private messages and non‑online communications [1] [4].
2. Geographic and data gaps that complicate a single national number
Investigations based on FOI requests reveal wide force‑level variation and gaps: at least six forces failed to respond or provided inadequate data (including Police Scotland), which means national totals compiled from partial returns understate both regional differences and possibly the overall count [3]. Local FOIs illustrate the problem of classification — West Yorkshire reported 24,703 arrests across certain harassment and malicious communications offence codes in a requested period, but only 1,533 records explicitly referenced platform keywords and 23,171 required manual review to determine if social media was involved [7]. Other forces similarly told requesters that extracting “social media” arrests requires manual chart review of offence narratives, so cross‑force aggregation involves assumptions [8].
3. Prosecutions, convictions and the “chill” narrative: the numbers diverge
Civil‑liberties commentators and EU parliamentary questions stress a chilling effect because arrest counts are headline‑grabbing, but prosecutions and custodial sentences are far rarer: reporting and fact‑checks show convictions and immediate imprisonment under these provisions have been low relative to arrests — for example, a 2024 year had only 137 immediate imprisonments referenced in reviews of the data — and the House of Lords library notes convictions have declined even as arrests rose [5] [4]. Myth‑checking outlets and analysts emphasise that many arrests do not result in charges or convictions and that the offences cover a mixture of trivial and serious cases, including some domestic abuse‑related or other serious communications [5] [4].
4. How reporting choices shape public perception
Different outlets frame the same custody and FOI material in divergent ways: some present the 12,000 figure as proof of mass censorship and “30 arrests a day” [6] [9], while fact‑checkers and police FOI responses point to classification limits, low conviction rates, and context such as post‑incident policing that inflate counts of communication‑related arrests [7] [8] [5]. Media summaries and parliamentary questions have amplified the number into policy debates about outdated laws and free‑speech impacts, but those debates must reckon with the inconsistent recording practices and incomplete responses from several forces [3] [4].
5. Bottom line and reporting limits
The most defensible, sourced answer is that reporting based on custody and FOI data puts arrests related to communications offences at roughly 12,000 in 2023 — commonly rendered as “about 30 arrests a day” — but that this number is an imperfect proxy for arrests solely for social media posts because the offence codes are broad, some forces did not supply full data, many arrests do not lead to prosecution or imprisonment, and local records often require manual review to confirm social‑media involvement [1] [2] [3] [7] [5] [4]. Where sources do not provide disaggregated, force‑by‑force, platform‑verified counts, definitive precision is not possible from the available reporting [8].